Managing Uncertainty in Scots Pine Range-Wide Adaptation Under Climate Change

Forests provide important ecosystem services and renewable materials. Yet, under a future climate, optimal conditions will likely shift outside the current range for some tree species. This will challenge the persistence of populations to rely on inherent plasticity and genetic diversity to acclimate or adapt to future uncertain conditions. An opportunity to study such processes is offered by Scots pine (Pinus sylvestris L.), a forest tree with a large distribution range including populations locally adapted to a wide variety of environments, which hinders a range-wide assessment of the species to climate change. Here we evaluate tree height growth uncertainty of Scots pine marginal populations in Spain and the Nordic countries linked to their genetic adaptation promoted by different climatic drivers. Our aims are to: (i) review the main climatic drivers of Scots pine adaptation across its range; (ii) undertake provenance-based modeling and prediction of tree height under current and future climate scenarios including four representative concentration pathways (RCPs) and five general circulation models (GCMs) at two extremes of its climatic niche; (iii) estimate uncertainty in population tree height linked to the main drivers of local adaptation that may change among RCPs and GCMs in the Nordic countries and Spain. Our models revealed that tree height adaptation is mostly driven by drought in Spain and by photoperiod in the Nordic countries, whereas the literature review also highlighted temperature as a climatic driver for the Nordic region. Model predictions for the Nordic countries showed an overall increase in tree height but with high uncertainty in magnitude depending on the RCPs and GCMs whereas predictions for Spain showed tree height to be maintained in the north and reduced in the south, but with similar magnitudes among RCPs and GCMs. Both models predicted tree height outside the data range used to develop the models (extrapolation). Predictions using higher emission RCPs resulted in larger extrapolated areas, constituting a further source of uncertainty. An expanded network of Scots pine field trials throughout Europe, facilitated by data collection and international research collaboration, would limit the need for uncertain predictions based on extrapolation